PDF documents can be encrypted using a variety of industry-standard schemes which have increased in complexity and security over the years, starting with PDF version 1.1. The PDF standard provides, in addition, a general mechanism for encapsulating third-party encryption and security policies.

Encryption applies, with a few exceptions, to streams and strings in the file, but does not encrypt numbers or other PDF data types, nor does it encrypt the file as a whole. Thus, the document’s object structure remains visible to applications without the need for decryption, but the substantive content of the document is safeguarded.

The more modern PDF encryption methods allow the file’s XMP metadata stream (XML Metadata) to be left unencrypted so it may be extracted and read by programs which don’t know how to open encrypted PDF files, or if the password is not known.